If Flaubert were alive today, he would be taking a special interest in Western policy on the Balkans. He was always fascinated by a certain kind of accepted wisdom, which shades off into platitudes, clichés, and expressions of sheer stupidity--what he lovingly described as la bêtise. An updated version of his Dictionary of Received Ideas would have to include several new entries derived from Western policymakers during the Bosnian war: "Balkan people: full of ancient ethnic hatreds. Cannot stop fighting one another." "NATO air strikes: completely ineffective without the deployment of hundreds of thousands of NATO ground troops." "Arming the victims: creates a level killing field. Only prolongs the war", and so on.

More recent events would have added a couple of new entries: "Kosovo, autonomy of: must be restored." "Kosovo, independence of: dangerous and destabilizing; would lead to new Balkan war." These two received ideas are constantly affirmed by our politicians and diplomats; the more they are repeated, the less often anyone pauses to question their truth. How could a policy assumption be wrong, when the foreign ministry of every major power in the West is agreed about it? The Bosnian experience suggests that the answer to that question is: very easily. Some serious thinking is needed about the possibility of independence as a long-term solution for Kosovo. If, as I believe, the foreign policy establishment has got this issue completely wrong, the consequences, in terms of Balkan instability and costly Western involvement--to say nothing of the lives of thousands of the local inhabitants--could be severe.

Already, the West's insistence that autonomy is the only solution has generated problems, both for Western diplomacy and for the Kosovo Albanians. Despite the self-congratulatory spin that Western governments put on Richard Holbrooke's October agreement with Slobodan Milosevic, it is clear that major concessions were made to the Yugoslav president. (Perhaps the most important was the abandonment of plans to have NATO-controlled observers backed up by NATO firepower; instead, the unarmed observers are controlled by the OSCE, a notoriously toothless, amorphous, and politically manipulable body.) Holbrooke was in fact in a very weak negotiating position. The message his political masters were sending to Milosevic was: "We shall attack your forces in Kosovo, going to war, in effect, on behalf of the Albanians against you--and then, when we have defeated your army, we shall turn round to the Albanians and tell them to go back under your rule, with a little regional autonomy to keep them happy." Milosevic must have known that this was illogical, and therefore he must also have known that the threat of military action could be heavily discounted.

Now that the deal has gone through, however, the illogicality is simply transferred to the West's dealings with the Kosovo Albanians. For more than six months American diplomats were preoccupied with getting the Albanians to form a united negotiating front. After the Holbrooke-Milosevic agreement was signed, the diplomats' aim has been to persuade those Albanians to negotiate for autonomy, and nothing more. But since the vast majority of Albanians in Kosovo want independence (having voted massively for it in an unofficial referendum as long ago as 1991), any local politician who signs up to mere autonomy now will be discredited, and perhaps even targeted in the first stirrings of a potential civil war. By insisting on a commitment to autonomy, Western diplomats will polarize Kosovan politics and undermine precisely those moderate Kosovo Albanian politicians whose role they most need to strengthen. It looks like a new application of the principle of "divide and rule": the West gets to divide the Kosovo Albanians, and Milosevic gets to rule them.

The way out of these immediate problems, and the way toward a genuine long-term settlement, lies in re-thinking, from first principles, the accepted arguments on autonomy and independence. These can be divided broadly into two categories: arguments about the intrinsic justifiability of independence, and arguments about its consequences. Let us take the intrinsic arguments first.

The main claim here is that Kosovo simply has no right, in constitutional or international law, to independence. The outside world has recognized (in, for example, the wording of Security Council Resolution 1199) that Kosovo forms part of the territory of a sovereign Yugoslav state; and as the diplomats never tire of repeating, the West is not in favor of changes to international borders. But these objections are precisely the ones that were made in 1991, when Slovenia and Croatia demanded independence. Eventually Western governments recognized those countries, having discovered that this involved not so much a change of borders as a change in the status of existing borders: the lines on the map remained the same, but their status was upgraded from republican to national.

Could Kosovo qualify for the same treatment? The answer, in terms of constitutional and international law, is that it could--and, indeed, that it should have been offered independence when the old Yugoslavia broke up in 1991-92. Under the Yugoslav constitution of 1974 Kosovo was equivalent in most ways to Slovenia, Croatia, and the other republics. True, its position--as an "autonomous province"--was not identical to theirs: in theory it had a dual status, being defined both as a component of the republic of Serbia and as a component of the federal Yugoslavia. But in practice it exercised the same powers as a republic, having its own parliament, high courts, central bank, police service, and territorial defense force; it was formally defined (from 1968 onwards) as part of the federal system, and it was represented directly--not via the Republic of Serbia--at the federal level. By all normal criteria of constitutional analysis, Kosovo was primarily a federal unit, and only very secondarily a component of Serbia.

In 1991 the European Community set up a committee of jurists, the Badinter Commission, to advise it on the break-up of Yugoslavia. The commission's key finding was that the whole federal system was in a process of "dissolution." In other words, what happened when Slovenia and Croatia became independent was not secession, not the falling away of a few branches from a continuing trunk; rather, the whole federal state dissolved into its constituent units. (The present-day "Yugoslavia" is not the continuation of the old Yugoslavia, but a new state, formed by the coming together of two units, Serbia and Montenegro.) Unfortunately, the Badinter Commission never said which units were the constituent ones, and Western governments simply made a policy decision to regard only the six republics as such--thus treating Kosovo as a wholly owned subsidiary of Serbia. Possibly they were influenced by the fact that, by this stage, Milosevic had already stripped away Kosovo's autonomous powers. But if Serbia's right to rule Kosovo is to be based on the mere fact that Milosevic had downgraded its status just before the break-up of Yugoslavia, it will rest on very shaky foundations, as the relevant constitutional changes were pushed through under extreme duress, with tanks in the streets and war planes roaring overhead.

The other intrinsic argument against independence for Kosovo is historical, not legal. Most Western diplomats seem to believe that Kosovo is an essential part of historic Serbian state territory, so that to remove it would be as bizarre as separating Yorkshire from England. This argument too is false.

Kosovo was not, as Serbs claim, the "birthplace" or "cradle" of the Serb nation, and it came under Serb rule for only the last part of the medieval period. Since then it has been excluded from any Serb or Yugoslav state for more than 400 out of the last 500 years. It was conquered (but not legally annexed) by Serbia in 1912, against the wishes of the local Albanian majority population, and it became part of a Yugoslav kingdom (not a Serbian one) after 1918. In other words, out of the entire span of modern history, Kosovo has been ruled from Belgrade for less than a single lifetime.

Of course it is true that the national mythology of Serbia--a mythology developed largely by nineteenth-century ideologists--sets great store by the historic importance of Kosovo, thanks to the site of the famous battle of 1389 and the presence of some important medieval monasteries, including the Patriarchate. But modern political geography cannot be determined by old battlefields, however symbolically charged they may be by the defeats incurred at them; if that were so, France would claim Waterloo, and Germany Stalingrad. Similarly, if modern borders had to bow to religious history, Kiev would be part of Russia and Istanbul part of Greece. Any independence deal for Kosovo would naturally have to include guarantees on the protection of cultural and religious sites; but that is a separate issue, and not such a hard one to resolve.

Aside from those intrinsic arguments, the Western diplomats also argue against independence for Kosovo on the grounds that it would set risky precedents or have dangerous consequences. A common claim is that if Kosovo gained independence, the Serb-ruled half of Bosnia, Republika Srpska, would also be entitled to break away from Bosnia. As Warren Zimmermann recently noted in these pages, "U.S. officials are particularly worried that Western acceptance of an independent 'Kosova' would destroy the Dayton agreement on Bosnia, which is based on integration, not separation" (Summer 1998).
But those officials are making a completely false parallel between the two cases. As explained above, Kosovo's independent statehood would be based on the fact that it--just like Bosnia--had been a unit of the old federal Yugoslavia; Republika Srpska never was such a unit, and indeed was granted legal status for the first time only in 1995, on the strict condition that it remain part of a sovereign Bosnian state. For most of modern history the territory of Republika Srpska has been an integral part of a Bosnian entity, whereas Kosovo has been legally attached to a Serbian entity only for the last fifty-three years.
The other argument involving precedents or consequences is about Macedonia, which has its own large Albanian minority. It is said that independence for Kosovo would encourage the Macedonian Albanians to carve off a territory of their own from the Macedonian state. In fact, the leading Albanian politicians in Macedonia make no linkage between independence for Kosovo, which they support, and a carve-up of Macedonia, which they do not want. One obvious reason why they do not want it is that more than 200,000 Albanians live in the capital, Skopje, which would certainly be left in the Slav half of any partitioned Macedonia.

But there is a different and real danger. A long, simmering conflict in Kosovo would gradually radicalize the Albanians of Macedonia, as their young men crossed the mountains to fight. Some of them would return home imbued with the wild rhetoric of "Greater Albania", which certainly exists in some branches of the Kosovo Liberation Army. Such radicalization would undermine the responsible political leadership that represents the Macedonian Albanians today; eventually, fighting could develop in Macedonia too. And the cause of this radicalization process--a long, simmering conflict in Kosovo--is precisely what Western policy guarantees when it denies to the Kosovars the one thing, independence, for which they are still determined to fight. Thus Western policy, which aims above all at preventing the destabilization of Macedonia, will create precisely the outcome it most fears.

What, then, can be done? Independence cannot come immediately to Kosovo: that would be too much of a shock to Serb pride, and would provoke a violent response. In the very long term, however, Kosovo will certainly be separated from Serbia; even some Serb nationalists concede this, when they compare birth rates and calculate that Albanians will outnumber Serbs in the whole of Serbia by the mid-twenty-first century. The solution, then, must lie in the medium-term--something along the lines of the settlement that ended the war in Chechnya, with a long interim period of autonomy leading finally to full self-determination. Conditionality could be built into such an agreement: to qualify for the eventual move to independence the autonomous Kosovo would have to satisfy key conditions, such as respecting the rights of the Serb minority and abandoning any territorial ambitions outside the present Kosovan borders. Such a solution would restore authority to the moderate Albanian political leaders, drawing support back toward them and away from the hardliners in the Kosovo Liberation Army. The continuation of the West's present policy on the other hand, far from solving Kosovo's problems, will only make them--and those of the whole Balkan region--far more lethally insoluble in the future.